惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

爱范儿
爱范儿
T
The Blog of Author Tim Ferriss
G
Google Developers Blog
博客园_首页
博客园 - 【当耐特】
量子位
S
SegmentFault 最新的问题
B
Blog RSS Feed
酷 壳 – CoolShell
酷 壳 – CoolShell
V
Visual Studio Blog
T
Tailwind CSS Blog
阮一峰的网络日志
阮一峰的网络日志
V
V2EX
Y
Y Combinator Blog
博客园 - 聂微东
The Cloudflare Blog
小众软件
小众软件
J
Java Code Geeks
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
月光博客
月光博客
H
Help Net Security
Jina AI
Jina AI
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
宝玉的分享
宝玉的分享

Hacker News: Show HN

PurrrrrFocus: Pomodoro Timer App - App Store Workflow Engine — Multi-Step Orchestration for Bun RapidPhoto: Pro Photo Editor App - App Store GitHub - DheerG/swarms: Achieve extraordinary results with claude code across a variety of tasks SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Show HN: VCoding – A 5 MB native Windows IDE with no dynamic dependencies Show HN: LLMs don't hallucinate because they're bad at math, it's the format GitHub - Agent-FM/agentfm-core: AgentFM is a peer-to-peer network that turns everyday computers into a decentralized AI supercomputer. AgentFM lets you run massive AI workloads directly across a global mesh of idle CPUs and GPUs. Show HN: Tracking Top US Science Olympiad Alumni over Last 25 Years GitHub - Potarix/agent-hub: One place to talk to all your agents Show HN: Runtime security for AI agents(injection,tool abuse, data exfiltration) GitHub - dubeyKartikay/lazyspotify: Terminal Spotify client for macOS and Linux GitHub - the-banana-tool/king-louie: Easy to use GUI Personal AI Assistant. Win/Linux/Mac. Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission GitHub - basteez/jsf-autoreload: maven plugin to enable hot reload on jsf projects uvm32/hosts/host-gdbstub at main · ringtailsoftware/uvm32 GitHub - labsai/EDDI: Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus. GitHub - glitchnsec/fortyone-oss: AI Executive Assistant Platform Quickstart | Alien GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. GitHub - ocrbase-hq/ocrbase: 📄 PDF/IMG ->.MD/JSON Document OCR API for PaddleOCR and GLMOCR. Self-hostable. GitHub - impactjo/home-memory: MCP server that lets your AI assistant remember everything about your home. GitHub - Sets88/dbcls: DbCls is a powerful terminal database client that supports various databases GitHub - neptun2000/heor-agent-mcp GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh RollQuation: Math Puzzles - Apps on Google Play GitHub - dropbox/witchcraft Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis GitHub - opentalon/opentalon: OpenTalon is an open-source platform built from the ground up in Go as a robust alternative to OpenClaw LinkedIn™ 职位抓取工具 - Chrome 应用商店
ModelHub — DeepSeek API: 97% Cheaper Than GPT-5.5
Eolab · 2026-06-08 · via Hacker News: Show HN

🔥 Product Hunt Launch June 11

45 Chinese AI Models.
One Key. 97% Cheaper.

DeepSeek V4, GLM-4, Doubao, Kimi — all through one account.
Same AI tier as GPT-5.5. No Chinese phone. Global payment.

✅ $5 free credit · No credit card · Visa / MC / PayPal

🎉 Your API Key is ready

Copy it and start using it right now.

Quick start: Replace api_key in your OpenAI SDK with this Key, change base_url to https://api.modelhub-api.com/v1

All systems operational 99.9% uptime 45 models online 100,000M+ tokens served today

Visa · Mastercard · PayPal

Paddle · PCI Level 1 certified · GDPR compliant

Available Models

DeepSeek V4 Flash — $0.15/M DeepSeek R1 — $0.55/M Qwen 3 — $0.10/M GLM-4 — $0.20/M Doubao — $0.12/M Kimi — $0.18/M +39 more

1

Get your free API key

Enter your email above. Your key appears instantly. No Chinese phone. No credit card.

// Your key looks like this: mh-sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

2

Choose how you want to use it

Two ways. Pick the one that works for you:

A

🌐 Use our Web Chat 推荐

No download, no settings — just open modelhub-api.com/chat in your browser and type. Free 1,000 requests included.

B

📦 Download a chat app

2. Settings → OpenAI

3. Enter API Key + URL + Model

Done. Free 1,000 requests included. Upgrade to $15/month. See pricing →

# Easiest — Environment variables (works everywhere):
export OPENAI_API_KEY="mh-sk-xxx"
export OPENAI_BASE_URL="https://modelhub-api.com/v1"

# Or — One line change in your code:
client = OpenAI(
    api_key="mh-sk-xxx",
    base_url="https://modelhub-api.com/v1"
)

# Or — Quick test via curl (Mac/Linux):
curl -s https://modelhub-api.com/v1/chat/completions \
  -H "Authorization: Bearer mh-sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Hello!"}]}'

# Windows (PowerShell):
$body = @{model="deepseek-v4-flash";messages=@(@{role="user";content="Hello!"})} | ConvertTo-Json
Invoke-RestMethod -Uri https://modelhub-api.com/v1/chat/completions \
  -Method POST -Headers @{Authorization="Bearer mh-sk-xxx"} -Body $body

Starter

$15

/ month

60M tokens (~300k pages)

DeepSeek V4 Flash · 1 key

Get Started →

🔥 Launch Special

Pro

$65

/ month

280M tokens (~1.4M pages)

All models · 5 keys · Priority support

Subscribe →

Use code PHLAUNCH50 for 50% off

Enterprise

Custom

Unlimited

Dedicated infra · 24/7 support

Contact Sales

Third-Party Benchmarks

Model Overall Reasoning Coding Math

GPT-5.5 93 92 94 91

Claude 4 90 88 92 87

DeepSeek V4 89 90 88 91

Gemini 2.5 86 84 87 82

DeepSeek V4 math score equals GPT-5.5 (91) and trails by just 4-6 points in other categories — at 97% lower cost.

Is the AI quality as good as GPT?

DeepSeek V4 Flash scores competitively with GPT-5.5 and Claude Sonnet 4 on independent benchmarks. Same tier, fraction of the cost.

Do I need a Chinese phone number?

No. Email signup only. Global payment via Visa/Mastercard/PayPal. No Chinese requirements at all.

Can I use this without downloading anything?

Yes! Just use Terminal (Mac) or Command Prompt (Windows). Copy the curl command from the Developer tab and paste it. The AI responds right there.

How do you protect my data and payments?

All traffic is SSL/TLS encrypted. Payments are processed by Paddle — a globally trusted payment processor. We are GDPR compliant. We never store your prompts or use them for training. Full privacy policy →

Who runs ModelHub?

We are a dedicated team of engineers and AI researchers with backgrounds from MIT, HKUST, Tsinghua, and USTC. Founded by Xiao Zhe (ex-infrastructure engineer), we are building the bridge between world-class Chinese AI models and global developers.

Email

ModelHub API - 45+ AI models, 43x cheaper than OpenAI